Trust and Reliance in XAI -- Distinguishing Between Attitudinal and Behavioral Measures

TL;DR

This position paper addresses the ambiguity in how transparency in explainable AI affects trust by distinguishing between attitudinal trust (a subjective psychological construct) and behavioral reliance (an observable action). It argues that inconsistent measurement practices underlie mixed empirical findings and provides a theoretical basis for separating the two concepts, drawing on established definitions and models of trust and reliance. The authors discuss measurement approaches (subjective scales vs. objective metrics like weight of advice) and emphasize probabilistic relationships between trust and reliance, proposing that future XAI research should measure both constructs and ground interpretations in robust theoretical frameworks. The work highlights practical implications for evaluating transparency in AI systems, aiming to improve validity, comparability, and the explanatory power of XAI studies.

Abstract

Trust is often cited as an essential criterion for the effective use and real-world deployment of AI. Researchers argue that AI should be more transparent to increase trust, making transparency one of the main goals of XAI. Nevertheless, empirical research on this topic is inconclusive regarding the effect of transparency on trust. An explanation for this ambiguity could be that trust is operationalized differently within XAI. In this position paper, we advocate for a clear distinction between behavioral (objective) measures of reliance and attitudinal (subjective) measures of trust. However, researchers sometimes appear to use behavioral measures when intending to capture trust, although attitudinal measures would be more appropriate. Based on past research, we emphasize that there are sound theoretical reasons to keep trust and reliance separate. Properly distinguishing these two concepts provides a more comprehensive understanding of how transparency affects trust and reliance, benefiting future XAI research.